Health and Social Welfare Review Vol.22 No.2, pp.134-174
This study aims to estimate the effect of tobacco tax increases on smoking behavior and initiation among youth by using the data collected from the National Longitudinal Study of Adolescent Health (Add Health) in 1995 and 1996. In addition, the author measured state-specific anti-smoking sentiment based on the data collected in September 1995 and May 1996 through the Current Population Survey, Tobacco Use Supplement. These measures are used as additional control variables in standard cross-section models of youth smoking behavior. For an analysis of smoking initiation among youth, the author used retrospective information on the age of initial cigarette smoking among students and estimated hazard models with state fixed effects. When no control was exerted on unobservable state-specific anti-smoking sentiment, price effects were negative and significant for both the cross-sectional models and the hazard models. In the smoking participation model, the price elasticity was -0.5384 for the 1995 cross-section and -0.4225 for the 1996 cross-section, a similar level as those seen in previous cross-sectional studies on youth smoking. When the state specific anti-smoking sentiment was controlled, however, estimated price effects were positive and insignificant. Findings from this study imply that the price effects found in previous cross-sectional studies could have been overestimated because they failed to control the state specific anti-smoking sentiment which is correlated with the cigarette price of the state. Also, the findings suggest that anti-smoking sentiment is a strong influence on youth smoking behaviors. It may be possible to design prevention policies that take advantage of this effect.
Table Of Contents
1. Introduction 2. Background 3. Cross-sectional Analysis on Youth Current Smoking and ever Smoking Behaviors 4. Hazard Analysis on Youth Onset Smoking Behaviors 5. Results 6. Conclusions References Summary